Center for Security Policy Studies

Alliances, Alignments, and the Future of Strategic Partnerships

CSPS Fellows Jordan Cohen and Connor Monie discuss the viability of formal alliances, and why visions of their demise are overblown.

In September 2019, Donald Trump publicly stated that he and Israeli Prime Minister Benjamin Netanyahu discussed a mutual defense pact. Ironically, while the U.S. State Department lists Israel as a “Major Non-NATO Ally,” no formal alliance agreement actually exists. The result is that Washington is not bound to come to Israeli aid if the latter enters a conflict.

This puts Israel in a category of countries that receive strong signals of support from the United States, but without a formally written alliance agreement. Foremost are Israel and Taiwan, while Saudi Arabia, the United Arab Emirates, and Mexico all receive large amounts of weapons and coalition aid, signaling a smaller commitment.

On September 25, 2020, George Mason University’s Center for Security Policy Studies debated the durability of the formal alliance. Given the risks of multiple, rival great powers in a world where allies and adversaries cooperate, should formal defense commitments die in lieu of short-term commitments?

Formal alliances should remain an integral component of states’ security policies for several reasons. First, the criticisms levied by alliance detractors rest on shaky empirical ground. Second, formal treaty arrangements offer substantial benefits during peacetime, from credibility value to military interoperability, intelligence sharing and joint training opportunities. Third, the alternative to formal alliances – liquid, informal, ad hoc alignments – remain conceptually imprecise, making relative comparisons ineffectual.

Why are Formal Alliances “Dying”?

The traditional argument for formalizing alliances is that it reduces problems associated with the alliance dilemma: that countries want their allies to join them in war but do not want to join a war outside of their own interests simply because an ally is involved. Formalization lowers the risks of abandonment because official treaties add reliability costs to the equation. In other words, as the cost of not showing up for an ally increases, the probability that states renege on commitments decreases.

On the other hand, many practitioners warn that formalization increases abandonment and entrapment risks. For example, alliance with Montenegro offers the US little in the way of fighting Russia, but a formalized agreement does not mean that the U.S. will risk Washington D.C. in exchange for Podgorica. To those who argue for the death of alliances, these formalized agreements do not improve the likelihood of an ally aiding the U.S. in war, but they also increase the likelihood that the U.S. will become engaged in a conflict where it has only limited interests.

Thus, the main benefit of informalized alignments is that they do not come with any requirement to fight war, nor do they exist absent a partnership. Aligned states can fight in coalitions together, as the case in Iraq and Afghanistan, but they do so without risking unnecessary conflict on behalf of the United States.

Whither Formal Alliances, Long-Live Formal Alliances:

While fears of an amplified alliance dilemma are worth discussing, they are largely based on poor evidence and analysis. There are several reasons for this. Firstly, there is little empirical evidence defending the argument that formal agreements increase risks present in the alliance dilemma. This is because these formal agreements are usually made after risk-analysis. Therefore, the United States would simply not sign a formal agreement with a country determined to be a risky ally. Research by Michael Beckley confirms this by examining the loopholes the United States includes in alliance agreements to avoid entanglement into dangerous wars that do not further US interest. Additionally, LeVeck and Narang find that countries who violate agreements tend to lose potential  allies, raising the costs of abandoning allies. Alliance agreements are only formed when risks in the alliance dilemma can successfully be avoided.

Second, viewing formal commitments as only important during wartime is problematic. The grand majority of “alliance years” occur during peacetime and the largest alliance since the Second World War has never actually fought together in combat as an alliance. Examining alliances only in the framework of war ignores the majority of alliance-related costs and benefits from formal alliances, especially examining NATO. Beginning one’s analysis at the onset of war excludes the myriad benefits alliances offer to states that seek to avoid war and deepen military interoperability with allies. For example, my research at Strategic Studies Quarterly finds that alliances are strongly correlated with weapons sales, partially because of the strategic benefits and trust stemming from written, formalized alliances. Ultimately, the “death of alliances” framework needs to account for the difference in peacetime benefits like interoperability training and weapons sales rather than strictly looking at wartime alliances.

Finally, the deeper, non-empirical problem is that the “death of alliances” thesis is a game of whack-a-mole. By not defining its terms, proponents can selectively point to individual successes or failures of agreements and claim them as actionable evidence. If informal alliances include any basic strategic partnership, the United States is aligned with 169 different countries. This definitional ambiguity allows proponents of “the death of alliances” thesis to select on the dependent variable. Cases where “alignments” are strong and have replaced strategic partnerships become the positive cases, whereas non-aligned states become the null cases. This fallacious reasoning makes it impossible for anyone to “disprove” their argument. Absent a clearer definition of alignment with explicit bounds, the “death of alliances” thesis is opportunistic and undeveloped.

None of these three issues mean that alignments are useless or that formal alliances are infallible, but that the nature and benefits of alignments are too underdeveloped to argue for the “death of alliances.” Moreover, it is a lofty task to argue for the end of alliances that have existed since the end of the Second World War. This is made all the more difficult by stating they will be replaced by vague, non-binding, and innocuous institutions that, by themselves, offer unclear benefits compared to formalized alliances. Collective defense treaties continue to supply security value that informal groupings, as currently defined, are incapable of providing. Consequently, better definitional reasoning is required before one can say Middle Eastern coalitions will overtake institutions like NATO.


Jordan Cohenis a political science PhD student at George Mason University’s Schar School of Policy and Government. Additionally, he holds an M.A. in Middle East and Islamic Studies. His work has been featured in the National Interest and Jadaliyya. His research focuses on the Arab Gulf, international alliances, and oil politics.

Connor Monie is a PhD student in political science at the Schar School of Policy and Government. He holds a bachelor’s degree in government and international politics with a concentration in international relations and a master’s degree in international security, both from George Mason University. His research interests center on interstate alliances, including alliance formation, management politics, and wedge dynamics. 

Photo can be found here